Methods and apparatus for analyte concentration monitoring using harmonic relationships

Abstract

Continuous glucose monitoring (CGM) may include applying a periodic excitation signal via an electrode of a CGM sensor to human interstitial fluid to drive an oxidation/reduction reaction, and measuring the current through the electrode. In some embodiments, the measured current is sampled and digitized, and various harmonics of the excitation signal's fundamental frequency are extracted. A set of relationships of at least two harmonics each is generated from the spectral amplitudes of a set of pairs, triplets, etc., of the harmonics, and the set of relationships is mapped to a glucose concentration such as based on the contents of a harmonic relationship database having a pre-existing set of harmonic relationships and glucose concentrations to which those sets of harmonic relationships correspond, for example. Numerous other embodiments are provided.

What is claimed is: 
     
       1. A continuous glucose monitoring (CGM) system, comprising:
 a CGM sensor configured for insertion into a region of interstitial fluid in a user; 
 a first electronic circuitry configured to couple to the CGM sensor and configured to be removably attached to an external surface of the user, wherein the first electronic circuitry includes a periodic excitation signal generator configured to couple to the CGM sensor, a current sensor configured to couple to the CGM sensor, and a sampling circuit configured to couple to the current sensor, the sampling circuit configured to output sampled time-domain data; 
 a second electronic circuitry coupled to the first electronic circuitry, wherein the second electronic circuitry is configured to extract a predetermined number of harmonics from the sampled time-domain data, generate a set of harmonic relationships based on the extracted harmonics, and determine a glucose level; and 
 a harmonic relationship database having a plurality of sets of harmonic relationships, wherein each set of harmonic relationships in the plurality of sets of harmonic relationships is associated with a known analyte concentration, 
 wherein the second electronic circuitry determines the glucose level based on a comparison between the plurality of sets of harmonic relationships of the harmonic relationship database and the generated set of harmonic relationships based on the extracted harmonics. 
 
     
     
       2. The CGM system of  claim 1 , wherein the first electronic circuitry and the second electronic circuitry are disposed in a wearable portion of the CGM system. 
     
     
       3. The CGM system of  claim 1 , wherein the first electronic circuitry is disposed in a wearable portion of the CGM system, and the second electronic circuitry is communicatively coupled to the first electronic circuitry. 
     
     
       4. The CGM system of  claim 3 , wherein the second electronic circuitry is disposed in a portable portion of the CGM system separate from the wearable portion. 
     
     
       5. The CGM system of  claim 1 , wherein the current sensor is configured to provide a voltage signal, and the sampling circuit is further configured to sample the voltage signal of the current sensor at a sampling rate greater than a fundamental frequency of a periodic excitation signal output by the periodic excitation signal generator by a factor between 50 and 400. 
     
     
       6. The CGM system of  claim 1 , wherein the sampled time-domain data is in a digital format. 
     
     
       7. The CGM system of  claim 1 , wherein the second electronic circuitry is configured to extract the predetermined number of harmonics from the sampled time-domain data by transforming the sampled time-domain data to frequency-domain data. 
     
     
       8. The CGM system of  claim 1 , wherein the second electronic circuitry includes Fast Fourier Transform circuitry. 
     
     
       9. The CGM system of  claim 1 , wherein the second electronic circuitry includes a processor, and a memory having instructions stored therein, coupled to the processor, wherein the instructions, when executed by the processor, cause the processor to:
 extract the predetermined number of harmonics from the sampled time-domain data. 
 
     
     
       10. The CGM system of  claim 9 , wherein the instructions, when executed by the processor, further cause the processor to:
 perform a transform operation including one or more of a Fourier transform, a Discrete Fourier Transform, a Fast Fourier Transform, and a Goertzel Transform. 
 
     
     
       11. The CGM system of  claim 1 , wherein the first electronic circuitry is configured to employ the periodic excitation signal generator to produce a periodic voltage signal having a frequency selected based at least in part on an approximate glucose level of the interstitial fluid. 
     
     
       12. A method of continuous glucose monitoring (CGM), comprising:
 generating, by a periodic excitation signal generator, a periodic excitation signal having an amplitude and a fundamental frequency; 
 applying the periodic excitation signal to an electrode of a CGM sensor, 
 wherein the fundamental frequency of the periodic excitation signal applied to the electrode is based at least in part on an approximate concentration of glucose in a glucose-containing fluid, 
 wherein the approximate concentration of glucose in the glucose-containing fluid is based on an expected glucose concentration, 
 wherein the expected glucose concentration is based on at least one of: a typical observed analyte concentration for a patient, previous glucose concentration levels measured for the patient, time of day the periodic excitation signal is applied to the electrode, when the patient last ate prior to when the periodic excitation signal is applied to the electrode, or what the patient last ate prior to when the periodic excitation signal is applied to the electrode; 
 sensing, by a current sensor circuit, a current through the CGM sensor to produce a measured current signal; 
 sampling, by a sampling circuit, the measured current signal at a sampling rate, for a period of time, at a bit resolution, to produce a set of time-domain sample data; 
 transforming the set of time-domain sample data to a set of frequency-domain data, wherein the set of frequency-domain data includes at least a strength of each one of a predetermined number of harmonics of the fundamental frequency; 
 generating a set of harmonic relationships based on the strength of each of the predetermined number of harmonics; and 
 determining a glucose level based on the set of harmonic relationships, 
 wherein an accuracy of determining the glucose level is increased at least in part on basing the fundamental frequency of the periodic excitation signal applied to the electrode on the approximate concentration of glucose. 
 
     
     
       13. The method of  claim 12 , wherein applying the periodic excitation signal comprises:
 applying the periodic excitation signal by a potentiostat. 
 
     
     
       14. The method of  claim 12 , wherein the periodic excitation signal is sinusoidal. 
     
     
       15. The method of  claim 12 , wherein the periodic excitation signal is triangular. 
     
     
       16. The method of  claim 12 , wherein the fundamental frequency is between 0.1 Hz and 10 Hz. 
     
     
       17. The method of  claim 12 , wherein the sampling rate is between 10 samples per second and 1000 samples per second. 
     
     
       18. The method of  claim 12 , wherein the set of frequency-domain data further includes a phase angle of each one of the predetermined number of harmonics of the fundamental frequency. 
     
     
       19. The method of  claim 12 , wherein sensing comprises:
 passing the current from the CGM sensor through a resistor having a precision of in a range of 0.1% to 1%. 
 
     
     
       20. The method of  claim 12 , wherein transforming the set of time-domain sample data to the set of frequency-domain data comprises:
 performing a Goertzel transform operation. 
 
     
     
       21. The method of  claim 12 , wherein transforming the set of time-domain sample data to the set of frequency-domain data comprises:
 performing a transform operation including one or more of a Fourier transform, a Discrete Fourier Transform, a Fast Fourier Transform, and a Goertzel Transform. 
 
     
     
       22. The method of  claim 12 , wherein generating the periodic excitation signal includes producing the periodic excitation signal having the fundamental frequency selected based at least in part on an approximate glucose level of an interstitial fluid.